摘要
针对公交乘客人流量统计准确度不高的问题,提出一种基于HOG和SVM的人流量统计算法。首先采用机器学习的方法,提取人头部的HOG特征,将SVM作为学习训练方法,得到关于人头的线性目标分类模型的分类器,成功检测出人头;其次通过数据关联,将Camshift算法作为人头目标跟踪算法,并利用tracking-by-detection机制,实现多目标跟踪,稳定地捕获人头目标的运动轨迹;最后对轨迹分析,判断目标是否越过设定的计数线,从而完成对公交乘客人流量的自动计数。实验表明该算法统计准确率有明显提高,且误检率较低,特别是在白天光照条件较好时,能够实现人流量的有效计数。
Aiming at the low statistical accuracy of passenger flow,this paper proposes a bus passenger flow calculation algorithm based on HOG and SVM.First,machine learning method is used to extract the HOG feature of the passenger heads;SVM is used as the training method,and the classifier for the linear model of target classification is obtained,then the heads are detected successfully.Next,Camshift is used as the head tracking algorithm;through data correlation and the tracking-by-detection mechanism,the multiple target tracking is achieved and the head motion trajectory of passenger target is captured stably.At last,the trajectory is analyzed and the automatic counting of bus passenger flow is realized.The experiment results show that the statistical accuracy of the proposed algorithm is improved effectively;especially during the daytime with good illustration,the effective counting of the passenger flow is achieved and the inward and outward passenger counting can be realized.
出处
《仪器仪表学报》
EI
CAS
CSCD
北大核心
2015年第2期446-452,共7页
Chinese Journal of Scientific Instrument